Development of a vaccine that is effective against the worldwide HIV pandemic has, to date, been unsuccessful. However, the ultimate feasibility of a useful vaccine is indicated by the fact that some individuals infected with HIV develop broadly neutralizing antibodies against the HIV viral envelope glycoprotein (Env) that provide protection against HIV infection in experimental systems. The current application describes a proposal to use random mutagenesis and screening approaches to provide better understanding of the antigenic properties of the envelope glycoprotein and to develop forms of Env with improved immunogenic properties. Based on the idea that poor affinity of Env for germline precursors to neutralizing antibodies is a major limitation on the ability of different forms of Env to elicit protective immune responses, a major goal of the project is the identification of variant forms of Env with the elevated affinity for binding such precursors. To facilitate mutagenesis and screening, the gp140 external segment of Env will be expressed at the surface of the baker's yeast Saccharomyces cerevisiae. This system was chosen based on the advantages of this yeast as a eukaryotic host for expression of heterologous proteins, because of the demonstrated usefulness of yeast for vaccine production, and because of the tools available for large-scale random mutagenesis and screening of surface-displayed proteins in yeast. The initial aim of the project is to optimize gp140 expression constructs to achieve useful levels of expression of protein exhibiting authentic antibody and receptor-binding capability at the yeast cell surface. Once these parameters are established, an initial round of mutagenesis and screening will be used for detailed mapping of the sequence determinants of epitopes for selected neutralizing antibodies, using Fluorescence Activated Cell Sorting (FACS) to identify Env variants with reduced affinity for the antibodies. In addition, the ability of yeast-expressed Env to bind to germline precursors to broadly neutralizing antibodies will be evaluated. In cases where such affinity is low (expected to be the majority of cases), libraries of randomly altered forms of Env will be screened by FACS to identify forms of the glycoprotein exhibiting enhanced affinity for the germline antibodies. If necessary to achieve an initial improvement in affinity, either single examples or libraries of partially matured forms of the neutralizing antibodies will e used for screening, followed by further iterations of screening to obtain Env variants exhibiting elevated affinities for actual germline antibodies. Candidate forms of Env exhibiting enhanced affinity for germline antibodies will be used to immunize rabbits. The resulting sera will be evaluated for their binding specificities and tested for the ability to neutralize diverse strains f HIV in an in vitro system. Overall, the proposal uses a new technology and approach for enhancing the immunogenicity of HIV Env that provides a complementary alternative to current structure-based design.